QUICK FACTS
- The City of Ottawa operates a diverse fleet of roughly 2,800 vehicles used in a wide range of services including garbage collection, snow removal, ambulance and fire response.
- The resulting model produced by Bronson for this project is a multi-variable model which forecasts annual maintenance costs based on vehicle data and characteristics where a relevant correlation was identified. These variables include age, annual usage, purchase price and using department.
“Bronson gathered 6-years of historical maintenance data and uploaded it to an Alteryx workflow built with Automated Machine Learning tools to analyze annual vehicle maintenance costs.”
– Phil Cormier, CA, CPA, Senior Consultant, Bronson Consulting Group
Project Description
Bronson was hired by City of Ottawa, Fleet Services to develop Alteryx workflows to forecast annual maintenance costs and develop a 10-year capital replacement plan for their motor vehicle fleet.
Business Challenge
The City of Ottawa required predictive analytics to anticipate how future vehicle maintenance costs and other related expenses would be impacted by an aging vehicle fleet. This analysis would assist the city in determining the optimal replacement timing for their vehicles in order to minimize overall operational costs.
Our Solution and Outcome
Bronson gathered 6-years worth of historical maintenance data for the municipal fleet and loaded this data into an Alteryx workflow built with Automated Machine Learning tools to identify relevant correlations and associations relating to annual vehicle maintenance costs. Data was normalized for outliers and anomalies, and Alteryx was used to identify the optimal forecasting model, including the relevant predictive fields to be referenced by that model.
The model was then used to predict annual maintenance costs for the fleet. These maintenance cost forecasts were then used to assist in planning optimal vehicle replacement timing over a 10-year forecast period. The model includes an optimal replacement plan and a capacity to override optimal replacements as required in response to resource constraints or other operational priorities.
The model also includes calculations to reflect the impact of transitioning towards green fleet options such as hybrid and electric vehicles.
